Towards Online Data Mining System for Enterprises

Data Mining System, Knowledge Discovery, Data Stream, OLAP

As the amount of generated and stored data in enterprises increases, the significance of fast analyzing of this data rises. This paper introduces data mining system designed for high performance analyses of very large data sets, and presents its principles. The system supports processing of data stored in relational databases and data warehouses as well as processing of data streams, and discovering knowledge from these sources with data mining algorithms. To update the set of installed algorithms the system does not need a restart, so high availability can be achieved. Data analytic tasks are defined in a  programming language of the Microsoft .NET platform with libraries provided by the system. Thus, experienced users are not limited by graphical designers and their features, and are able to create complex intelligent analytic tasks. For storing and querying results a special storage system is outlined.